{"apiVersion":"v1","taxonomyVersion":1,"generatedAt":"2026-10-29T15:43:46.116Z","data":{"recordId":"wsa_m6nt2kx6b3bn","score":0.3,"confidence":"low","capped":false,"capReason":null,"sources":[{"n":1,"url":"https://mistral.ai/news/announcing-mistral-7b/","publisher":"Mistral","authority":"owner-controlled","verification":"verified","lastRetrievedAt":"2026-10-28T00:39:38.860Z","detectedPublishedAt":"2023-09-27T08:00:00.000Z","rights":"original"}],"claims":[{"id":"443acab2-180e-46c9-a511-5590bd15f8a0","kind":"capability","statement":"Mistral 7B has 7.3 billion parameters.","citations":[{"source":1,"supportProbability":0.39,"supports":false,"quote":null}]},{"id":"2e55b98f-4d59-4984-a0bf-a70333236599","kind":"capability","statement":"Mistral reports that Mistral 7B outperforms Llama 2 13B on all evaluated benchmarks.","citations":[{"source":1,"supportProbability":0.2,"supports":false,"quote":{"text":"Mistral 7B significantly outperforms Llama 2 13B on all metrics, and is on par with Llama 34B (since Llama 2 34B was not released, we report results on Llama 34B).","rights":"original"}}]},{"id":"0c01c5ae-0ab9-4f32-9f45-aefb3e0e6a0a","kind":"capability","statement":"Mistral 7B uses grouped-query attention for faster inference.","citations":[{"source":1,"supportProbability":0.26,"supports":false,"quote":{"text":"Uses Grouped-query attention (GQA) for faster inference","rights":"original"}}]},{"id":"9dc3d13b-ff84-40ac-9b13-1676cb6f2293","kind":"capability","statement":"Mistral reports a 2x speed improvement for sequence length 16k with a 4k sliding window.","citations":[{"source":1,"supportProbability":0.31,"supports":false,"quote":{"text":"In practice, changes made to FlashAttention and xFormers yield a 2x speed improvement for sequence length of 16k with a window of 4k.","rights":"original"}}]},{"id":"c73dcfe9-5dfd-4789-aa5c-18acdfdfa695","kind":"availability","statement":"The model is available to download and use locally.","citations":[{"source":1,"supportProbability":0.13,"supports":false,"quote":{"text":"Download it and use it anywhere (including locally) with our reference implementation ,","rights":"original"}}]},{"id":"b223d487-b2b1-4c19-b85d-e9ac8fa70adb","kind":"deployment","statement":"The model can be deployed on AWS, GCP, or Azure.","citations":[{"source":1,"supportProbability":0.21,"supports":false,"quote":{"text":"We thank the teams of HuggingFace, AWS, GCP, Azure ML for their intense help in making our model compatible everywhere.","rights":"original"}}]}]},"nextCursor":null,"license":{"catalog":{"name":"CC BY 4.0","spdx":"CC-BY-4.0","url":"https://creativecommons.org/licenses/by/4.0/","attribution":"wowsuchai","attributionUrl":"https://wowsuchai.withglint.app","covers":["catalog metadata","classifications","generated summaries","feed event metadata"]},"excluded":{"covers":["source quotations","linked source content"],"terms":"Not licensed by wowsuchai. Source quotations and linked source content keep their original rights."}}}